Dynamic Discrete Choice Model

187 views
Skip to first unread message

Parastoo Jabbari

unread,
Mar 9, 2021, 3:59:32 AM3/9/21
to Biogeme
Dear Michel,

I'm building a Dynamic Discrete Choice Model, and I was wondering if there are any examples of using Biogeme for such models.

If not, do you think Biogeme can handle DDCM? Is there any precautions for using Biogeme for building such models?

Best,
Parastoo Jabbari

Bierlaire Michel

unread,
Mar 9, 2021, 5:16:49 AM3/9/21
to jab...@uw.edu, Bierlaire Michel, Biogeme
You may be interested in the recent paper that we have published on this topic:
https://transp-or.epfl.ch/documents/technicalReports/BierFrejHill21.pdf

It depends on what you need by “DDCM”.

Biogeme is designed to handle panel data. You’ll find several examples online.
http://biogeme.epfl.ch/examples/swissmetro/12panel.py

It easy to include past choices as explanatory variables. You need to deal with endogeneity, though.

If you want to use dynamic programming to model the “forward” behavior, this is not implemented in Biogeme.

Particle filtering is not implemented either.
> --
> You received this message because you are subscribed to the Google Groups "Biogeme" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to biogeme+u...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/biogeme/b2598dfc-ddac-4208-b3d2-dd26cb8ba4c7n%40googlegroups.com.

Antonin

unread,
Mar 14, 2021, 9:42:31 AM3/14/21
to Biogeme
We have a published a paper in 2016 about a dynamic model of location choice using panel data in the Journal of Choice Modelling: https://doi.org/10.1016/j.jocm.2016.04.003 It corrects for endogeneity due to lagged dependent variable and agent effect, in particular implementing a correction method by Wooldridge. The model specification files for Pythonbiogeme are available as additional material or directly under https://doi.org/10.5281/zenodo.619128 It is not written in PandasBiogeme, the latest version, but maybe it is still useful to you.
Reply all
Reply to author
Forward
0 new messages